From the Research Universities edition of November 2, 2026
Brown and Vanderbilt receive $732,574 from Schmidt Sciences to build DeepAndes Workbench, an agent-queried AI survey tool for Andean archaeology
Brown University (Parker VanValkenburgh); Vanderbilt University (Steven Wernke, Yuankai Huo); Oak Ridge National Laboratory; Schmidt Sciences · University humanities and social-science research (AI-assisted archaeological survey) · United States, with collaborators in Peru, Chile and Australia
- Publisher
- Brown University
- Original publication
- 2 November 2026 (Brown University news)
- Source retrieved
- 2026-11-03
What happened
Brown associate professor Parker VanValkenburgh and Vanderbilt researchers Steven Wernke and Yuankai Huo received $732,574 from Schmidt Sciences for "Scaling Archaeological Vision: Imperial Landscapes of the Andes through AI-Assisted Survey." They will build DeepAndes Workbench, which lets archaeologists query an agent with images and verbal prompts to find specific feature types, with expert feedback used to improve the models. It builds on GeoPACHA, their satellite-imagery and AI mapping platform. The team includes about 15 collaborators across the US, Peru, Chile and Australia plus Oak Ridge National Laboratory researchers; Schmidt Sciences also provides in-kind supercomputing. VanValkenburgh described it as a tool to extend researchers' vision, not an all-seeing eye.
Why it matters
Shows AI-for-research funding and national-lab compute reaching the humanities, a segment research computing groups often under-serve. It is also an example of philanthropic in-kind compute supplementing campus resources.
Evidence and measured results
Read the full release. This is an award announcement; no model performance, accuracy or user data are given. Expected outputs (continental-scale maps) are aspirations.
Limitations and uncertainty
University award announcement (institutional self-report, classed as operator claim); no results yet. Humanities use case with distinctive data; transfer to other fields is limited.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-02; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Humanities teams winning AI grants need imagery storage, compute and data governance that campus services rarely tailor to them.
- Stakeholders
- Research computing director
- Dean of humanities or social sciences
- Sponsored programs office
- Principal investigators
- Discovery questions
- How do you support externally funded humanities AI projects?
- Where will large imagery datasets live?
- Who governs sensitive site-location data?
- Value hypothesis
- Planned support for storage, compute and data agreements may let humanities AI projects start faster.
- Potential engagement
- A review of campus support for one externally funded humanities AI project.
- Unsupported claims
- The tool detects features accurately
- Campus compute is required
- Continental-scale maps will be delivered
Pre-sales engineering
Role takeaway
- Fit
- Fits geospatial computer-vision projects with expert annotation and external compute.
- Architecture and integration
- Satellite imagery store, vision models, an agent query interface and expert feedback loop, with philanthropic supercomputing.
- Prerequisites
- Imagery storage
- GPU access
- Annotation tooling
- Constraints
- Partner-country data rights
- Sensitive site locations
- Unreported model performance
- Security
- Restrict precise site coordinates to authorized researchers to reduce looting risk.
- Proof of value
- Precision and recall on expert-labelled areas
- Expert feedback turnaround
- Access control test on site data
Delivery
Role takeaway
- Work
- Data management, storage, multi-site compute access, annotation workflows and collaborator onboarding.
- Dependencies
- Schmidt Sciences compute
- ORNL collaboration
- International partners
- Ownership
- Principal investigators with research computing and data services support.
- Skills and adoption
- Archaeologists learn to query and correct the agent; staff learn geospatial AI support.
- Governance checkpoints
- Partner data agreements
- Site-location release review
- Grant reporting
- Acceptance criteria
- Documented detection accuracy on labelled regions (proposal)
- Collaborators onboarded
- Data plan approved
- Risks
- Model errors misdirect fieldwork
- Site data leakage
- Dependence on in-kind compute
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Image-search-style agent over satellite imagery with expert feedback loops, backed by external supercomputing from a philanthropy and ORNL collaboration.
Governance
Who approves, reviews and stays accountable for outcomes?
International collaboration on cultural heritage data raises questions about site-location sensitivity and partner-country data rights, not addressed.
Security and privacy
What data, permissions and controls need testing?
Precise archaeological site maps can enable looting; release controls are not discussed.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Expert users provide feedback that trains the models, adding annotation work for scholars.
Procurement
What should contracts, pricing and exit terms secure?
Compute comes in kind from Schmidt Sciences rather than campus purchase.
Operating model
Which teams own the service once it runs?
Multi-institution, international team with external compute; campus research computing role is unclear.
Publication history
- 2026-11-02Research Universities · Issue 553 resources
Stable resource ID: brown-vanderbilt-deepandes-workbench-schmidt-ai-archaeology-20261102